We have recently spent some time discussing The Cannabis Inventory Monster, and we have in the past discussed Cash Flow from Operations. Today we will merge these topics to discuss inventory management in a growth cycle, and what happens to Cash Flow from Operations during that cycle.

We have seen posters on various forums putting up the raw figures for Cashflow from Operations over multiple quarters and making grand proclamations.  Not one of these “analysts” has commented on the composition of the figure.  This is as novice level of analysis as looking solely at Net Income to gauge the health of the operations of a company, or simply reading the company’s earnings release for due diligence. Just like Net Income requires the onion to be peeled to determine what operationally happened during a period, so does the Cash Flow from Operations figure. Taking these numbers on whole can lead to faulty analysis.

I think the risks that these novices espouse are very real, especially for cash short companies. But sales growth takes time to play out, as does getting a cultivation facility from actual to projected volume. The risks could be realized if sales do not expand, and in some cases companies need sales growth to expand considerably if they intend to ramp production at greater volumes than presently harvested.

A very simple example helps illustrate where the novice thinker “steps on a rake”:

• If your A/R drop from $80 million to $40 million, all else being equal, your Cash Flow from Operations improves by $40 million.

Those touting novice level analysis would rejoice as Cash Flow from Operations improved by $40 million. But at the same time your sales might just have dropped by 50% as A/R is a function of sales volume and speed of collection, and the provinces are not cutting their payment terms by half. So, your Cash Flow from Operations improved but your business might have just taken a gut punch.

Let’s get to it….

Cash Flow from Operations is found on the “Statement of Cash Flows” in the financial statements and it is always stated as year to date review. The Statement of Cash Flows takes the dynamic that the Income Statement measures over a quarterly period, couples that with the cash deltas of other items from the balance sheet which are snapshots. The idea is to determine all the elements that contributed to a change in cash over the period in review. Cash Flow from Operations is one element. The others being: Cash Provided/Used in Financing Activities and Cash Provided/Used in Investing Activities. (Anecdotally: I do not think you can recreate the Statement of Cash Flows provided from the Income Statement and Balance Sheet without greater disclosure from the issuer.  This always makes me nervous to rely on information I cannot reproduce from the source documents, as we have seen management has not been at its best in using estimates and with disclosure and even the auditors have missed items and required whistleblowers to correct audits.)

Cash Flow from Operations consists of two main components:

1. OPEX Cash burn/generation and
2. Changes is non-cash based working capital items (eg. Non- cash items in the “Current” categories of the Balance Sheet).

Opex Burn includes most of the elements that adjusted Earnings Before Interest, Tax and Depreciation (“EBITDA”) adds back

• Opex Burn adds back like EBITDA non-cash items:  depreciation, gain on Bios, fair value gains, share based compensation…) from Net Income,
• Opex Burn does not add back cash items: Taxes, Interest and Transaction Costs (the latter in “adjusted” EBITDA) are not added back as they are paid in cash.

This OPEX Burn is the “cash” that operations in the period has used (“burned”)/generated in the period under review.

Example: Aphria Statement of Cash Flow February 29, 2020

OPEX Burn = ($124,367) less ($102,367) = ($21,426) is OPEX Burn for the nine months ended February 29, 2020.  The first three individual Q’s OPEX Burn is as follows: Q1 ($14,690), Q2 ($3,943), and Q3 ($ 2,523).

The other component is Non-Cash based working capital (the -$102,367 above), and REMEMBER it is the CHANGE in the asset/liability type that is recorded in Cash Flow from Operations, which largely consists of:

• Trade Accounts Receivable – largely 60 day owed from provincial buyers and other customers in cannabis. Medical direct (versus Veterans Affairs reimbursement which will have an A/R) is cash based, as such no A/R are generated. Aphria and Tilray’s non cannabis segments, Cronos CBD business also generate A/R although I am not sure on the tenor of trade terms availed.

• Inventory and Biological Assets stripped of IFRS voodoo. So ONLY the CASH COST COMPONENT is measured.

• Trade Accounts Payable – and as the bulk of value from what a cultivators sell is the plant there isn’t a huge % of Cost of Goods Sold made up of third party supplies/inputs, accordingly A/P growth isn’t as correlated to sales growth as much as in other industries with a large stream of third party inputs (eg. Vehicle manufacturing). Packaging supplies would be in here. There would also be components with 2.0 vape hardware.

• Other cash based changes in current assets and current liabilities.

So, if you were an LP building inventory in hopes of sales increases this is the successful cadence that would lead to increases in non-cash working capital QoQ.

1. Inventory build (keep in mind harvested inventory takes 5-7 weeks to become “saleable” which prolongs the inventory build)
2. Inventory build PLUS A/R build (keep in mind adult use A/R take 60 days or 8 weeks to collect)
3. Inventory stabilizes and A/R builds
4. A/R stabilizes (assuming sales plateau)

These are natural steps in a growth business. Remember the Cash Flow from Operations is the CHANGE QoQ in Inventory and A/R. Once either plateaus the change from that component is $0. There is a natural ramp in use of cash for non-cash working capital, which is extended by harvests not being saleable the day of harvesting, that needs navigation by LPs.

In the above you will notice that there is 13-15 weeks (a full Q) between when inventory is harvested to saleable inventory to collection of the cash from provincial buyers. These are the vast majority of uses of cash from the non-cash working capital group. When I speak of “operating float” (or float) it means the amount of capital required to bridge (or “float”) a sales cycle.

So, if you are forecasting a $50 million increase in sales in a Q you need your non-cash working capital to go up at least $50 million to bridge/float the cycle (which is pulled from cash). BUT the other thing to keep in mind is you need an inventory of product larger than your current Q sales to service your customers, as you are not selling just one homogenous SKU. That means that $50 million invested in non-cash working capital to support a $50 million increase in sales is actually insufficient for the enterprise.

Inventory Management:

We have not seen the cannabis industry zero in on what the appropriate multiple of inventory needed to service sales should be. Should a company hold 6 months of inventory, 9 months or more? This will depend on both their anticipated sales growth and the depth of product SKUs. A more homogenous inventory (single grade commodity) requires less inventory to be held in reserve to cover sales than does an inventory with 100 SKUs. Some SKUs might require 3x quarterly sales carried in inventory as sales are still increasing, while other SKUs might need less.

Cannabis Inventory to Quarterly Cannabis Sales: Peer and Trend

In the automotive industry they have strived over decades to get to Just in Time (JIT) inventory, at least on the manufacturing side. There are cash and cost benefits to getting to JIT and it is considered an optimal state if it can be achieved, as it reduces the funds need for a business to operate (via reduced inventory) and it reduces all the expenses of carrying inventory (eg. storage, finance costs).

But whereas the auto manufacturing industry can pull inventory from suppliers based on specific vehicle manufacturing build schedules in cadence, the legal cannabis industry is flying by the seat of their pants. Provinces are getting better at ordering inventory. The lessons learned by over filling provincial inventory for 1.0 were muted in the 2.0 rollout. But the provinces are guessing at what the consumer wants and will repetitively purchase. And until consumer demand shows a solid floor by SKU, producers are using their best estimates stocking inventory volumes.

The problem we have seen in the cannabis industry in getting cultivation ramped is like getting an oil tanker underway. It takes a lot of effort to get it moving. But then turning, stopping or course correcting the oil tanker becomes the issue. The agility of these large cultivation facilities is not a strong suit, or at least no one has demonstrated agility yet. Agility is further limited by the length of time it takes to grow the product then the length of time post harvest for the product to become saleable.  A decision to change harvest today (aside from a destruction of crops) will not be recognized in saleable inventory for a quarters time at a minimum (and that’s assuming you have Mothers in house to feed the cultivation process).

Where a business can go off the rails in a growth cycle as it relates to Cash Flow from Operations is when the projected sales increase does not materialize. You have built inventory, and you likely have more incoming (the “oil tanker”), but sales did not occur to drain any of the prior increase in inventory.

We have seen this happen to Canopy, OGI, Hexo and Cronos. And all but OGI are also shovelling cash into the Opex Burn at a high rate as well.

Canopy Cash Flow from Operations: Trend

CGC increased harvests from 14,000 kgs in Dec 2018 Q0-3 above (cannabis inventory +$200 million at that date), to two Q’s of +40,000 kgs, then 30,000 kgs in Q0. Their highest Q of sales on record is 13,000 kgs. So, for the last three Q’s they have been harvesting 2-3 times in excess of sales, growing a large inventory relative to sales volume even larger.

Canopy also has a significant OPEX Burn issue to contend with.

As Canopy built out their inventory in Q0-3 and Q0-2 they invested significantly in non-cash working capital in inventory.  The issue they have had is during that span Sales fell, and just rebounded in Q0. As such inventory increased.  Amounts Receivable has fluctuated very little since March 31, 2019 with a tight range of $103-109 million.  Had their sales increased the $ used in Working Capital would have added to the deficit.

In one calendar year Canopy added 82,000 kgs in inventory (+6x their highest quarterly sales on record) to an inventory pile that started at +$200 million increasing to +$550 million – not including writedowns of +$70 million).

Aurora Cash Flow from Operations: Trend

Aurora has built inventory and we are essentially waiting to see if they can get on top of it and turn the inventory into cash via sales, while not building as much replacement inventory (as they need to conserve cash) or building it at the same percentage change as sales.

Aurora has a similar but lesser issue to Canopy on Opex Burn.

In Q0-1 Aurora added 29,000 KGS to inventory as their harvest hit a record 41,000 KGs against sales of 12,500 KGs.  Again in this Q Aurora sold $10 million in wholesale revenue.

The positive $ Used in Working Capital in Q0-2 was fabricated to some extent by wholesaling $20 million in wholesale biomass, of which an undisclosed portion went to RTI.  This had the impact of fluffing GM, EBITDA and Opex Burn by $12 million and lowering $ Used in Working Capital by close to $20 million.

Over the past two Q’s a reduction in A/R of $21 million has improved $ Used in Working Capital and Cash Flow from Operations.  But the $59 million decrease in quarterly sales over that time period isn’t anything to cheer about.

If you looked at the raw figures as the novices have been doing, you would not understand the machinations required to deliver the above results.

Aurora is essentially restarting Step 2 from above:  Inventory build PLUS A/R build (keep in mind adult use A/R take 60 days or 8 weeks to collect)

Aphria Cash Flow from Operation: Trend

Aphria lagged behind CGC, OGI, and Aurora in cultivation capacity and inventory build. They have worked their OPEX burn down nicely during that delay.

Aphria still has to manage what all the others have failed to do, and that is to avoid having inventory runaway on them IF sales do not keep pace. Aphria inventory management will be severely tested over the next two Q’s as Diamond harvests over a full Q (harvests should jump from 32,000 kgs to 44,000 kgs this Q). Part of the harvest increases expected in Q1-Q2 F2021 will go to feed new product platforms like edibles and beverages. They will likely build a reasonable inventory level to feed the retailers before they start selling. But if Aphria is successful in achieving their projected production capacity of 63,750 kgs per quarter, they will need outlets other than Canadian adult use and medical to absorb production unless the Canadian market explodes to higher volume.

Aphria has entered Step 2 from above: Inventory build PLUS A/R build (keep in mind adult use A/R take 60 days or 8 weeks to collect).  But this time they are doing it with self-produced inventory and phasing out new third party inventory purchases.

Inventory Management Strategies:

Wholesale revenue has been one strategy used to moderate and manage inventory by a number of LPs. The variability of wholesale revenue by its “one off” or “opportunistic” nature is not as preferred as medical revenue or adult use revenue (as long as it doesn’t get returned through provisions or rebates) in generating a solid base-load of cannabis revenue.

Wholesale Revenue: Trend and Peer

The Wholesale market for cannabis will likely develop over time to become more stable and likely at lower transaction volumes. But presently it is not a dependable outlet for product, as you can see by the relative lumpy Q0 to Q0-3 bars above.

In addition to wholesale revenue as an inventory management tool Aphria also recently used a sleeving strategy (buying 3rd party). Aphria used wholesale PURCHASES to help bridge the increase in salable inventory by allowing them to build sales in advance of the first wave of Diamond inventory coming to market.

Had they not… the Diamond inventory would have started hitting the books and they would have needed the incremental sales (had they not sleeved sales) to substantially increase from a standing start.

Example:

• Last Q, of the $45 million in Aphria’s medical and adult use sales, sleeved sales accounted for $20 million with $5 million in gross margin.
• So, this Q instead of building on their Adult Rec base of $45 million they would have been building off of a base of $25 million.
• If they projected to get to $65 million in the next Q in sales, they need “only” a $20 million increase (+44%) from $45 million versus a $40 million increase (+160%) from $25 million had they not “sleeved”.

The sleeved inventory is a short term bridge that makes the incremental sales increase more manageable. Think as the sleeving of inventory as Aphria getting a running start at when Diamond inventory became saleable. When they are no longer buying inventory, self produced inventory will step in to try and hold that market share generated by the sleeved inventory. This is the difference between a 100 meters sprint and a relay race where the baton is handed off on the fly.

The sleeved inventory also increased their inventory ramp, using Cash Flow from Operations.

(BTW: I have seen some folks on investing forums criticize this sleeving approach. “Why are they buying product when they have product in inventory?”. Which is pretty idiotic and very simplistic. If someone with a fistful of money wants to buy product from you, the worst thing you can do is say you do not have that particular product to sell.  It is better, if you have the capital to afford it, to purchase inventory and make half the gross margin you would have on your own product. You make money. You build the market share.)

We saw Aurora do something very similar before Sky came online; purchasing from third parties for sales. Albeit, Aurora did not spell it out to the level of disclosure Aphria has provided. Aurora did not segment the sales from sleeved inventory and gross margin, nor did they segment purchase inventory in their notes to financials to be able to track the product.  They also did not do this with the extracted RTI product that came back on the Biomass Boomerang.

Next Moves???

On the last CC Carl estimated that current Q “working capital” (Note: Not the same as “non-cash working capital” as it relates to Cash Flow from operations) ) draw on cash could be another $25-50 million in this Q. Last Q the draw was $51 million for non-cash working capital.

Looking at last Q’s inventory cash cost increase of $28 million and extrapolating to new harvest amount, the higher end of that range will be tested without increased sales. And without visibility to maturities or segmentation between cannabis and distribution in the A/R, which increased faster/more last Q than the sales increase (which means there could be a lump in A/R that comes down this Q), there could be further pressure on the upper end of the range. The GGB note is due in July 2020, so that will not help paydown working capital use. (Note: GGB noie would free up cash and would be neutral to working capital and non-cash working capital items. Note 2: the GGB note was accelerated and paid in May 2019)

Given the conference call was halfway through the Q, I am hoping Carl adjusted for current circumstances. But a late Q wholesale sale, like last Q on Feb 20, 2020 but without swap component, could be a wrench in those plans.

The nature of large-scale cultivation of cannabis (that oil tanker) seems to have a six-month window where, if cultivation is left unchecked or sales increases do not materialize, inventory can get out of control quickly.

Aurora might have throttled their grow last Q, given the 10,000 kg drop in harvest.  We will see if their “Sales to Harvest ratio” and “Harvest to Sales delta” improve in the March 31, 2020 Q with a full Q of 2.0 products revenue.

Aphria will have to make a similar consideration about throttling the grow after the present Q. (40,000 kgs has been achieved by both Aurora-once and Canopy-twice, with Aphria on track for May 30, 2020 Q.  And this level of inventory has never been even 50% sold within the Q, we do not have data yet on 2Q’s after a 40,000 KG harvest). This decision will have to be done as they build inventory for non-vape 2.0 products to launch in Q1-Q2 F21, which comes with it all the risks inherent in new launch sales uncertainty.

Without the covid issues affecting sales, I would be comfortable betting that a full Q of vape sales (last Q was $6 million and 510’s and Alberta were later in Q and they repriced vapes mid Q to be more competitive) could incrementally replace most, if not all, of the wholesale revenue from last Q. With covid issues I am not that confident.

In closing, Cash Flow from Operations can be a useful gauge of the cash being consumed in operations. But the raw number is not nearly as important as understand the intra-levering of its aggregate parts.

Managing growth of cultivation in the cannabis industry at the scale we are seeing has never been done before. We have seen a number of different companies’ approach scaling inventory and managing inventory differently. To date, each of these companies have written down millions in inventory.

Aurora and Aphria are now on the clock to see if they can generate sustainable sales increases and manage/absorb their inventory and cultivation levels. If they don’t, they will face a similar fate as CGC, CRON, OGI, TLRY and HEXO with inventory impairments and decisions about how and when to throttle the grow.

Based on Aphria disclosure last Q in the MDA harvests should increase by approximately 12,000 kgs this Q to 44,000 kgs. They sold 10,000 kgs adult and medical and 5,000 kgs wholesale last Q. They are entering a very critical cycle for inventory management. Aphria ended the Q with $163 million in cannabis inventory, being over $200 million next Q is more than likely, but because of IFRS voodoo the actual cash cost (usage) is much less. (Eg. last Q the cash cost of inventory grown by Aphria increased $28 million versus balance sheet inventory growth of $74 million. The balance was purchased inventory of $21 million and $25 million FVI). If Aphria cannot gain more sales they will be looking at their inventory to sales ratio jumping from under 3:1 to closer to 4:1 in the next quarter.

Aurora is coming off the Dec. 31, 2019 Q harvest of 31,000 kgs down QoQ from 41,000 kgs. They sold 9,500 kgs down QoQ from 12,500 kgs that had a chunk of wholesale. Aurora has $206 million in inventory. Aurora is battling a dwindling cash supply. They launched a full suite of 2.0 products which should help move inventory but will create an increase in A/R. Keeping inventory growth minimal by increasing the KGs sold QoQ is a must for them as they were already approaching 4:1 inventory to sales last Q.

It’s amusing that folks posting their free “analysis” on Cash Flow from Operations on other forums, analysis that neglects any discussion of the individual components of Cash Flow from Operations. Many are pouting and think they aren’t being taken seriously enough. Their flawed, or rather incomplete, analysis could be the reason they are not getting traction from folks that understand financial statement analysis.

Cash Flow from Operations is useful. But it’s more useful when you understand the components and their correlation to net sum.

Will Aurora and Aphria fall into the same inventory trap as their peers? They very well might. Covid has dampened store openings and likely foot traffic. Online sales saw some “pantry loading” at first but that softened.

Navigating the inventory management hill is steep for Aurora and Aphria, with Aurora more exposed due to their relatively higher inventory to sales at last Q and their dwindling cash position.

And once a likely throttling occurs, yields will decrease. Meaning less product generated/yielded to cover fixed Cost of Goods Sold, and yields will likely drop faster than the semi variable Cost of Goods Sold can be throttled. Meaning compression of Gross Margin… and yet another issue to navigate in this journey. You better hope your company can cover expenses even with compression in Gross Margin because that is likely the next danger filled step of evolution in the industry. 

The preceding is the opinion of the author and is in no way intended to be a recommendation to buy or sell any security or derivative. The author has a position in Aphria and will not start one or divest in the next five days. The writer does not hold, nor will not start, a position in any of the other companies listed above in the next five days.